Definition
What is CRO?
Conversion rate optimization, or CRO, is the discipline of increasing the share of visitors who complete the intended action on a site: a purchase, a quote request or a booking. It does not look for more traffic. It removes the friction that makes visitors who are already there give up, then tests what remains uncertain.
What CRO looks at on a site
CRO starts from one specific action: a paid order for an online store, a received quote request for a B2B or B2C business that sells on quotes. It then follows the path that leads there, page by page, and looks for the point where visitors drop off.
Three sources feed this work: usability problems already documented by research, measurement of real behaviour on your site, and controlled experimentation such as the A/B test.
Why CRO matters in a marketing budget
A gain obtained through CRO then applies to all the traffic that reaches the site, paid or free. Doubling traffic usually doubles the acquisition budget; improving the conversion rate is paid for once.
The order of the work matters for the same reason. Raising ad bids before fixing the page that receives the clicks can mean paying more for visits that go nowhere.
What CRO does not fix
CRO amplifies what already exists. It does not fix an offer the market does not want, a poorly positioned price, or traffic made up of people who never intended to buy. When the problem is upstream, the conversion rate can rise slightly while sales stay flat.
Conversion rate = actions counted ÷ visits × 100
CRO is judged on this rate. Compare the rate with your own history, never with an industry average.
Fictional example: a business gets 8,000 visits a month on its site and 160 quote requests, a rate of 2%. It shortens its form and fixes an error message that was blocking submissions on mobile. The following month, with the same traffic, it receives 200 requests: the rate rises to 2.5%.
We read CRO as the cheapest lever in an acquisition budget, on one condition: fix what is known before testing. Most of the initial gain comes from corrections that are already documented (fees shown late in checkout, a form that is too long, mandatory account creation). Testing those points means paying to rediscover them. Tests come afterwards, on what is specific to your market.
We also judge the result in sales and leads, not in rate alone: a rate that rises because the site counts downloads brings in nothing.
Not to be confused with
- Conversion rate
- The conversion rate is the measure: the share of visits that end in the action. CRO is the work that moves that measure up.
- A/B test
- The A/B test is one CRO tool, not CRO itself. It settles what research has not already settled.
- Website redesign
- A redesign rebuilds the whole site. CRO removes specific friction on the existing site, usually without a redesign.
Related concepts
Further reading
- CRO: what conversion rate optimization really fixes
- Good conversion rate: what it is, and for which type of site
- A/B testing: why most tests prove nothing
- Forms: the number of fields decides, not the number of steps
- Cart abandonment: what the Baymard Institute's 49 studies show
- Heatmaps: what they show, and what they do not prove
Related services
Frequently asked questions
Is CRO just running A/B tests?
No. Most of the initial gain comes from corrections already documented by research, which require no test. A/B tests come afterwards, to settle what is specific to your market.
Do I need a lot of traffic to do CRO?
Not to fix what is already known: a form that is too long or fees shown too late can be fixed without a test. An A/B test, on the other hand, needs enough visits and conversions to tell a real difference from chance.
